1. Volume 7, Issue 2, Pages 163-173 (February 2003)
A canine conditionally replicating adenovirus for evaluating oncolytic virotherapy in a syngeneic animal model 
Akseli Hemminki, Anna Kanerva, Eric J Kremer, Gerd J Bauerschmitz, Bruce F Smith, Bin Liu, Minghui Wang, Renee A Desmond, Anne Keriel, Brian Barnett, Henry J Baker, Gene P Siegal, David T Curiel 
Molecular Therapy 
Volume 7, Issue 2, Pages (February 2003)
DOI: /S (02)
Copyright © 2002 The American Society of Gene Therapy Terms and Conditions

2. FIG. 1
(A) Activity of the osteocalcin promoter (OC) in a plasmid construct on canine osteosarcoma cells (D22, CF11, D17), a previously reported positive line (ROS), or a human colon cancer line (LoVo). Plasmids featuring either the OC or the ubiquitously active cytomegalovirus immediate early promoter (CMV) were transfected into cells and luciferase activity (relative light units, RLU) was measured 72 h later. Error bars indicate standard deviation. (B) Adenoviruses with OC (AdEasy-OC-Luc-CMV-GFP) or CMV (AdEasy-CMV-Luc-CMV-GFP) controlling expression of luciferase were constructed. Cells were infected and luciferase was determined at 24 h. Results are presented as the RLU ratio of OC vs CMV. DK and DKCre are nonmalignant canine kidney cells and BTB are nonmalignant canine mammary epithelioid cells. (C) A primary canine osteosarcoma sample was obtained fresh from the operation and homogenized. The osteosarcoma cells were then infected with two different doses of AdEasy-OC-Luc-CMV-GFP (OC) and AdEasy-CMV-Luc-CMV-GFP (CMV), followed by luciferase detection at 24 h. Mock-infected cells are also displayed. (D) To investigate if CAV-2 can replicate in and cause oncolysis of canine osteosarcoma cells, a TCID50 titering assay was performed. Cells were infected with serial dilutions of virus and the development of CPE was followed over 10 days. In the higher dilutions, CPE was observed at ca. 1 week, suggesting effective infection, productive replication, and oncolysis with CAV-2.
Molecular Therapy 2003 7, DOI: ( /S (02) )
Copyright © 2002 The American Society of Gene Therapy Terms and Conditions

3. FIG. 2
Schema for construction of OC-CAVE1, a canine conditionally replicating adenovirus. A plasmid containing the full-length CAV-2 genome (pTG5412) was digested and religated to create pCAVE1. PCR-directed mutagenesis was employed to create a unique KpnI site for insertion of the OC promoter. Homologous recombination was performed in Escherichia coli cells, resulting in a plasmid containing the full-length OC-CAVE1 genome, which was released and propagated on DKCre cells. CMV-CAVE1 was constructed in a similar manner, but contained the CMV promoter instead of OC.
Molecular Therapy 2003 7, DOI: ( /S (02) )
Copyright © 2002 The American Society of Gene Therapy Terms and Conditions

4. FIG. 3
Crystal violet cell killing assay. Canine osteosarcoma cells (D22, CF11, D17) or nonmalignant cells (BTB, DK) were infected with an E1-deleted vector (CAVGFP), the novel canine conditionally replicating adenovirus featuring the osteocalcin promoter for controlling E1A expression (OC-CAVE1), the isogenic control virus with a ubiquitously expressed promoter (CMV-CAVE1), or wild-type CAV-2. The growth medium was changed every 2 days until almost complete cell killing was visually evident at the lowest dose for any virus. Each cell line was evaluated individually. Crystal violet staining was then performed to detect attached cells. Killing of canine osteosarcoma cells with CAV-2-based agents did not always cause detachment although cells were dead upon visual inspection.
Molecular Therapy 2003 7, DOI: ( /S (02) )
Copyright © 2002 The American Society of Gene Therapy Terms and Conditions

5. FIG. 4
MTS cell killing assay. Canine osteosarcoma cells (D22, CF11, D17) or nonmalignant cells (BTB, DK) were infected with an E1-deleted vector (CAVGFP), the novel canine conditionally replicating adenovirus featuring the osteocalcin promoter for controlling E1A expression (OC-CAVE1), the isogenic control virus with a ubiquitously expressed promoter (CMV-CAVE1), or wild-type CAV-2. The growth medium was changed every 2 days until almost complete cell killing was visually evident at the lowest dose for any virus. Each cell line was evaluated individually. Mitochondrial activity was then determined and compared to that of mock-infected cells to estimate the proportion of live cells. Bars indicate standard deviation.
Molecular Therapy 2003 7, DOI: ( /S (02) )
Copyright © 2002 The American Society of Gene Therapy Terms and Conditions

6. FIG. 5
Therapeutic efficacy of OC-CAVE1 in vivo. D22 canine osteosarcoma cells were injected into both flanks of nude mice (n = 5/group) followed by intratumoral injection of viruses. Tumor size was measured every 3 days. (A) 1 × 109 VP of the canine conditionally replicating adenovirus featuring the osteocalcin promoter for controlling E1A expression (OC-CAVE1), the isogenic control virus with a ubiquitously expressed promoter (CMV-CAVE1), wild-type CAV-2, or no virus (OptiMEM) was injected on 3 consecutive days. Tumors treated with OC-CAVE1 were smaller that tumors treated with OptiMEM (P < ). (B) A single 1 × 108 VP injection of OC-CAVE1, CMV-CAVE1, CAV-2, OptiMEM, or an E1-deleted vector (CAVGFP) was performed. OC-CAVE1 displayed therapeutic efficacy over OptiMEM (P < ), CAV-2 (P < ), CMV-CAVE1 (P = ), and CAVGFP (P < ). Bars indicate standard error.
Molecular Therapy 2003 7, DOI: ( /S (02) )
Copyright © 2002 The American Society of Gene Therapy Terms and Conditions